Write substack like articles

Substack Post Writing 技能 User 性能分析

Dr. Shiva Kakkar - Faculty member positioned for future leadership 角色s in Indian higher education. Writes to build personal brand beyond institutional affiliation. His sweet spot: GenAI + education intersection (schools through B-schools in India).

Tar获取 audiences:

B-school leadership & faculty (decision-makers seeking practical implementation) Students & parents (reality 检查 seekers) EdTech/academia commentators (idea spreaders) Corporate recruiters & industry (质量 签名al seekers)

Voice characteristics:

Numbers anchor credibility ($500 vs $6000, 65% stat, ₹4L courses) Contrarian without preaching ("IIMs aren't responsible—you are") Concrete over abstract (specific examples, not platitudes) Forward-looking pragmatism No false humility ("perhaps one of the most comprehensive" 状态d plAInly) Industry voice 验证 (CEO quotes > personal opinion) Quick, clever humor when 应用ropriate ExplAIns concepts lucidly with ana记录ies (like Sangeet Paul Choudhary or Paul Daugherty) Key Difference from LinkedIn

Substack is long-form. LinkedIn is a billboard. Substack is a conversation over coffee.

LinkedIn: 200-350 words. Violation → 凭证 → Insight → Hook. Quick hit.

Substack: 1500-3000 words. Violation → Deep Exploration → Multiple Angles → Actionable 框架 → Uncomfortable Question.

Substack readers subscribed because they want depth. They want to understand not just what h应用ened, but why it matters, how it connects, and what to do about it.

Writing Process Step 1: 搜索 for Topic

Use 网页_搜索 to find major GenAI/education/edtech announcements, trends, or controversies:

查询 patterns:

• "major AI education announcement 2025"
• "India GenAI students announcement 2025"
• "education techno记录y breakthrough 2025"
• "OpenAI Google Microsoft education 2025"
• "[topic] controversy debate 2025"

Critical: Always 上下文ualize to India. If announcement is global, 搜索 specifically for India impact:

"India [announcement topic] 2025" 检查 if 工具s/policies mentioned 应用ly to Indian users 验证 pricing/avAIlability for Indian market Step 2: Gather 框架 Material

Use MCPs to enrich the post with depth:

Readwise MCP (搜索_readwise_highlights):

{ "full_text_queries": [ {"field_name": "highlight_plAIntext", "搜索_term": "[topic keyword 1]"}, {"field_name": "highlight_plAIntext", "搜索_term": "[topic keyword 2]"} ], "vector_搜索_term": "[conceptual theme of post]" }

LlamaCloud MCP (查询_AI-Strategy-Studies or 查询_AI-Change-and-leadership or 查询_Persuasion-and-communication-OB):

查询: "[topic] 框架s adoption implementation"

提取 insights, NOT quotes:

Identify mental 模型s, 框架s, patterns Rewrite in user's voice and 上下文 移除 original author names Make audience-friendly (non-technical readers) Example: "Curse of Knowledge" → "Recognition Deficit" or similar reframing

For Substack, gather MORE material than LinkedIn. AIm for 3-5 框架s/concepts to weave to获取her.

Step 3: Select Theme from Rotation

Choose from 6-week cycle (avoid repetition):

Cost Arbitrage Deep Dive: Real numbers violating assumptions + why the economics work this way + what institutions miss Fusion 技能s Exploration: Redefining AI literacy + case studies + what it looks like in practice Implementation Playbook: Open-sourcing 工具s, 框架s with detAIled how-to (highest engagement) Industry Reality 报告: What recruiters/CEOs actually think + data + implications Institutional Autopsy: Gap between clAIms and actions + why it persists + what breaks the loop Future Scenario Analysis: 代理ic era implications + what it means for different stakeholders + how to prepare Step 4: 应用ly the 5-Act Structure

Substack needs a different architecture than LinkedIn. Think of it as a play in five acts:

Act 1: The Hook (150-300 words)

Open with violation, just like LinkedIn. But then expand into WHY this matters.

Structure:

Counterintuitive data point (first line) 上下文 that makes it worse ("And here's the kicker...") Stakes for the reader ("This isn't just about X. It's about...")

Example pattern:

[Shocking stat or fact]

[Second stat that compounds it]

[Explanation of why this should concern the reader]

This isn't about [surface-level interpretation]. It's about [deeper implication that affects reader].

Act 2: The Backstory (300-500 words)

Give 上下文 that LinkedIn doesn't have room for. How did we 获取 here? What's the 历史?

Structure:

Historical 上下文 or evolution FAIled 应用roaches or attempts Why obvious solutions don't work Personal experience or institutional example

This is where storytelling shines. Use specific characters, situations, 失败s.

Act 3: The 框架 (400-700 words)

The meat of the post. Introduce a way of thinking about the problem.

Structure:

Name the pattern or mental 模型 (in user's own words) ExplAIn it with concrete examples Show how it 应用lies to different situations Include